Abstract Cancer cells reprogram metabolic pathways to survive and proliferate in response to changes in their microenvironment. While oncogenic pathways sustain glycolytic metabolism to enhance survival during tumorigenesis, mitochondrial metabolism is significantly altered upon loss of extracellular matrix (ECM) contact and growth under anchorage-independent conditions. By applying metabolic flux analysis (MFA) to tumor spheroids, we have identified particular changes in serine, alanine, and sphingolipid metabolism that control tumor cell growth in such microenvironments. Many tumors amplify or overexpress phosphohydroxypyruvate dehydrogenase (PHGDH) and other enzymes within the serine synthesis pathway to support growth, though the specific mechanisms through which this pathway supports aggressive tumor growth remain unclear. Serine and alanine metabolism are linked via sphingolipid biosynthesis, where the enzyme serine palmitoyltransferase (SPT) produces toxic deoxysphingolipids (doxSLs) in the context of abundant alanine and low serine levels. These atypical doxSLs are produced at higher rates during anchorage-independent growth and compromise mitochondrial metabolism to mitigate spheroid growth. By modulating the production and availability of serine, alanine, and sphingolipids we can control in vitro and in vivo tumor growth. These findings provide a novel and unexplored mechanism through which serine deprivation limits cancer cell growth. This proposal aims to exploit the production of doxSLs in tumors by manipulating dietary amino acids and endogenous serine synthesis to mitigate tumor growth and metastasis. In Aim 1 we will apply MFA to characterize changes in mitochondrial and amino acid pathways during anchorage-independent growth. In Aim 2 we will quantify how sphingolipid biosynthesis is impacted during spheroid growth and determine why doxSL species are toxic to tumor cells. In Aim 3 we will design specifically formulated diets that mitigate tumor growth and metastasis by modulating doxSL production when administered alone or in combination with PHGDH inhibitors. We will also engineer the sphingolipid biosynthesis pathway in tumor cells to validate our central hypothesis. If successful, this proposal will define a novel mechanism through which serine deprivation limits tumor growth that can be exploited via dietary interventions and used to identify responsive tumor types in the clinic.